Network Uncertainty Informed Semantic Feature Selection for Visual SLAM
This addresses the challenge of efficient long-term localization for autonomous vehicles, though it is incremental as it builds on existing SLAM methods with new feature selection.
The authors tackled the problem of long-term visual SLAM by developing SIVO, a feature selection method that incorporates semantic segmentation and neural network uncertainty to reduce map size. The result was comparable performance to ORB_SLAM2 on the KITTI dataset while reducing map size by almost 70%.
In order to facilitate long-term localization using a visual simultaneous localization and mapping (SLAM) algorithm, careful feature selection can help ensure that reference points persist over long durations and the runtime and storage complexity of the algorithm remain consistent. We present SIVO (Semantically Informed Visual Odometry and Mapping), a novel information-theoretic feature selection method for visual SLAM which incorporates semantic segmentation and neural network uncertainty into the feature selection pipeline. Our algorithm selects points which provide the highest reduction in Shannon entropy between the entropy of the current state and the joint entropy of the state, given the addition of the new feature with the classification entropy of the feature from a Bayesian neural network. Each selected feature significantly reduces the uncertainty of the vehicle state and has been detected to be a static object (building, traffic sign, etc.) repeatedly with a high confidence. This selection strategy generates a sparse map which can facilitate long-term localization. The KITTI odometry dataset is used to evaluate our method, and we also compare our results against ORB_SLAM2. Overall, SIVO performs comparably to the baseline method while reducing the map size by almost 70%.